In Africa, children less than five years of age bear the greatest burden of HIV and malaria co-infection. Even with the use of daily trimethoprim-sulfamethoxazole (TS) prophylaxis and insecticide treated bednets (ITNs), malaria remains very common in HIV-infected children that live in areas with high malaria transmission intensity. The antiretroviral protease inhibitor (PI) lopinavir inhibits P. falciparum in vitro at levels lower that those achieved in serum with standard dosing of lopinavir/ritonavir (LPV/r). In this study, we will test the hypothesis that the use of Pi-based antiretroviral therapy (ART) will reduce the incidence of malaria compared to non-nucleoside reverse transcriptase (NNRTI)-based ART among HIV-infected children living in a high malaria transmission area of Uganda. The three specific aims of this study are: 1. To compare the incidence of malaria among HIV-infected children receiving PI vs. NNRTI-based ART 2. To compare the safety and tolerability of PI vs. NNRTI-based ART 3. To compare the virologic and immunologic efficacy of PI vs. NNRTI-based ART The study will be a randomized open label trial of 300 HIV-infected Ugandan children, 6 months to 5 years of age that meet Ugandan and World Health Organization criteria for ART initiation. The study will take place in Tororo, Uganda a district with high malaria transmission intensity. The primary study endpoint will be the incidence of symptomatic malaria. Secondary endpoints include the incidence of complicated malaria, prevalence of anemia and asymptomatic parasitemia, incidence of adverse events grade 2 or greater, CD4 cell recovery and prevalence of HIV RNA viral suppression. Children will be treated with either PI- or NNRTIbased ART and followed for 24 months. All children will receive TS prophylaxis and ITNs. All medical care will be provided at the study clinic, using well validated protocols for malaria diagnosis, treatment and followup. The study will be conducted by a team with extensive experience in the study of malaria and care of HIV-infected children. This trial will test a novel strategy for the prevention of malaria in one of Africa's most vulnerable populations. The targeted use of Pis as first line therapy for young HIV-infected children that reside in malaria endemic regions could significantly reduce the high morbidity and mortality from both diseases.